INDEX

Explanations

references to additional information or examples

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 EconPapers

-0.84

OGND

-0.74

⟬

-0.73

 CreateTagHelper

-0.69

ChildScrollView

-0.69

astify

-0.69

exitRule

-0.66

MLLoader

-0.66

jsonwebtoken

-0.64

 المعيارى

-0.61

POSITIVE LOGITS

etc

0.35

etc

0.34

 einiges

0.33

 całość

0.31

Anyhow

0.31

anyway

0.30

 francès

0.30

low

0.30

weed

0.30

0.29

Activations Density 0.036%

references to additional information or examples

No Comments

No Known Activations